Post on 22-Sep-2020
Size and Focus of a Venture Capitalist�s Portfolio
Paolo Fulghieri Merih Sevilir
University of North Carolina University of North Carolina
paolo_fulghieri@unc.edu merih_sevilir@unc.edu
October 30, 2006
We thank an anonymous referee, Robert McDonald (the editor), Michael Fishman, Eitan Goldman, Josh Lerner,
seminar and conference participants at the American University in DC, the University of Amsterdam, UNC, FRA
Conference in Las Vegas, RICAFE Conference in Turin, and the second FIRS Conference in Shanghai for very
useful comments. All errors are our own.
1
Abstract
In this paper we take a portfolio approach to analyze the investment strategy of a venture capitalist
(VC), and study the optimal size of a VC�s portfolio. We show that portfolio size and scope a¤ect the
incentives of both entrepreneurs to exert e¤ort and VCs to make start-up speci�c investment. A small
portfolio improves entrepreneurial incentives because it allows the VC to concentrate his limited human
capital on a smaller number of start-ups, adding more value to each start-up. In addition, by holding
a small portfolio, the VC commits not to extract higher rents from the entrepreneurs, with a positive
impact on their incentives. A large and focused portfolio is bene�cial for the VC because it allows the
VC to reallocate his limited resources and human capital from one start-up to another in case of a start-
up failure. Furthermore, a large and focused portfolio allows the VC to extract greater rents from the
start-ups because the VC can induce competition for his limited resources. We show that the VC �nds
it optimal to limit his portfolio size when start-ups have higher quality prospects ex ante, that is, when
providing strong entrepreneurial incentives is most valuable. The VC expands his portfolio size only when
start-up fundamentals are more moderate and only when he can form a su¢ ciently focused portfolio.
Finally we show that the VC may �nd it optimal to engage in portfolio management by divesting some of
his start-ups early since this strategy allows him to extract a greater surplus from the remaining start-ups
in his portfolio. Surprisingly, we �nd that under certain conditions, portfolio management, despite being
socially ine¢ cient ex post, improves ex ante social welfare by enlarging the set of economically viable
start-ups.
1 Introduction
Existing theoretical work on venture capital (VC) has so far concentrated on a VC�s investment in a
single entrepreneurial start-up, where the VC provides monetary and non-monetary resources to turn the
entrepreneur�s project idea into a viable business.1 However, VC funds typically invest in more than one
start-up at any given time, and engage in active portfolio management to maximize the return from their
investment. In recent research, Kaplan and Schoar (2005) �nd substantial heterogeneity in performance
across private equity funds of di¤erent size, and suggest that VCs with superior skills and greater human
capital can generate better results in their investments. They also �nd that better performing VC funds
grow proportionally slower and argue that better VCs may choose to stay small (by deliberately limiting
the amount of capital raised) to avoid dilution from allocating their limited amount of human capital over
a large number of start-ups.2
In this paper we take a portfolio approach to analyze the investment strategy of a venture capitalist
and investigate the optimal size of a VC�s portfolio. More speci�cally, we address the following questions:
What determines the size of a VC�s portfolio? What are the bene�ts and costs of having a small versus
a large portfolio? What are the strategic aspects of managing a portfolio of start-ups? Do VCs prefer
having a diversi�ed portfolio as opposed to a focused one? How do size and focus of a VC�s portfolio a¤ect
performance?
Our analysis shows that holding a small portfolio can be an optimal strategy for a VC even if the VC
has access to a large number of potentially pro�table start-ups in the economy. Our starting point is to
recognize that VC human capital is a scarce resource in the economy which cannot be augmented easily.
Even if the VC has the ability to raise unlimited amount of capital for a large number of start-ups, this
may not be the optimal strategy if the VC cannot back up his monetary investment by his human capital.
Spreading his human capital and diluting his value adding capability over a large number of start-ups may
a¤ect performance so adversely that the VC may �nd it optimal to limit the size of his portfolio.
Our analysis starts with the notion that both entrepreneurial e¤ort and VC human capital are essential
inputs for a given start-up�s success. In addition, a key feature of our analysis is to recognize that VC
1 Inderst, Mueller and Muennich (2006), Kanniainen and Keuschnigg (2003) and Bernile and Lyandres (2003) provide
notable exceptions.2For anecdotal evidence on the relevance of fund size see The Economist, which wrote on April 2, 2005 that �some venture
capital funds say they have turned away money from investors in order to keep fund sizes down to an amount that can be
managed responsibly�.
1
human capital is a �xed resource in limited supply, and cannot be expanded easily. The basic trade-o¤ we
investigate is that whether a given VC should concentrate all his human capital and resources on a small
number of start-ups, or spread them over a larger number of start-ups.
We show that the size and scope of the VC�s portfolio a¤ect both entrepreneurial incentives to exert
e¤ort and the VC�s incentives to make start-up speci�c investment. A small portfolio is bene�cial for
the VC for two di¤erent reasons. The �rst is that in a smaller portfolio the VC can add more value
to each start-up and, as a response, each entrepreneur �nds it optimal to exert higher e¤ort, improving
success potential of the start-up. The second bene�t of a small portfolio is that by investing in a small
number of start-ups, the VC limits his ability to induce competition among start-ups for his limited human
capital and resources. In other words, by holding a small portfolio, the VC commits not to exploit the
entrepreneurs by threatening to take resources away from one start-up and transferring them to another
one. This commitment proves bene�cial for ex-ante entrepreneurial incentives.
There are bene�ts associated with holding a large portfolio as well. The �rst is that having a larger
number of start-ups increases the VC�s ex-post bargaining advantage when the start-ups compete for
his limited human capital at a future project�s stage. Thus, increasing the number of start-ups in the
portfolio allows the VC to extract a higher surplus from each entrepreneur. The second bene�t of a large
portfolio is that it allows the VC to reallocate resources from one start-up to another in case one start-up
fails. We show that the magnitude of both bene�ts associated with a large portfolio becomes greater as
the relatedness of the start-ups in the portfolio increases, that it, as the VC�s ability to form a focused
portfolio increases. This follows from the fact that a more focused portfolio increases both the VC�s rent
extraction ability and his resource reallocation e¢ ciency.
Our main results hinge on the balance between the bene�ts and the costs of a small versus large
portfolio and the VC�s ability to form a focused portfolio. We �nd that a small portfolio is more desirable
when start-ups have a higher potential payo¤, lower risk and a lower level of relatedness. These are exactly
the conditions under which promoting strong entrepreneurial incentives outweighs the cost of a reduction
in the VC�s rent extraction ability and resource reallocation e¢ ciency. In contrast, when start-ups have a
lower expected return, higher risk and higher degree of relatedness, it becomes more desirable for the VC
to form a larger portfolio. Note that a larger portfolio weakens entrepreneurial incentives, but this proves
to be less costly for start-ups with lower expected returns and higher risks, since entrepreneurial e¤ort will
be lower in such start-ups even in a small portfolio.
Our model also highlights the value of active portfolio management, a common VC practice observed
2
in real life. We show that a VC with a large portfolio may �nd it optimal to divest one of his start-ups
early, even if the company�s early stage performance is positive. Early disposal of a start-up may result in
the early termination of an otherwise potentially viable venture, or in its sale to another VC fund, or in an
early Initial Public O¤er (IPO). This portfolio management strategy is desirable from the VC�s point of
view, since divesting a start-up allows the VC to add more value to the remaining start-ups in his portfolio
and to extract more surplus from them. We �nd that the strategy of early divestiture is optimal when
the start-ups in the portfolio have a high degree of relatedness and hence, when the VC has a focused
portfolio. We also show that the practice of portfolio management can increase ex ante social welfare by
enlarging the set of start-ups �nanced by the VC.
Our paper makes several novel contributions. This is the �rst paper, to our knowledge, which studies
the interaction of size and scope of a VC�s portfolio. We analyze the costs and bene�ts of a large versus
small portfolio as well a focused versus a diversi�ed portfolio. Our paper shows that a VC may prefer to
limit the size of his portfolio even if he has access to a large number of potentially pro�table start-ups.
Note that the VC�s desire to limit portfolio size is not the outcome of the assumption that the number of
good projects is limited, but it derives from the bene�t of providing entrepreneurs with stronger incentives.
In our model the VC may prefer to limit his portfolio size precisely because expanding portfolio size will
have a negative spillover e¤ect on the existing investments.3 Furthermore, we show that the VC will �nd
it desirable to have larger portfolios only when he can form a portfolio of su¢ ciently related start-ups, that
is, when he can e¢ ciently reallocate resources from one start-up to another. Since the ability to reallocate
resources proves to be most valuable for start-ups with high risk and failure rates, this implies that VCs
investing in high-tech and risky industries will be more likely to have larger portfolios.
Note that the contribution of our paper is not limited to VC investment only. Our paper also speaks
to the more general topic of the theory of the �rm by studying both project size and scope together and
adds to the literature on the theory of the �rm in terms of the optimal number of divisions as well as
their relatedness in a given �rm. Existing research on the theory of the �rm and internal capital markets
considers the advantages and disadvantages of �rms with a large number of divisions (see, for example,
Gertner, Scharftein and Stein, 1994), but is silent about the relatedness of divisions within a �rm and its
impact on optimal �rm size .
3Thus our paper provides a theoretical explanation for the observation in Kaplan and Schoar (2005) that �passing up
less pro�table (but potentially still positive NPV projects) could only be an optimal choice for the GP if there are negative
spillover e¤ects on the inframarginal deals from engaging in these investments,� (page 1822).
3
In an extension of our model, we analyze how changes in the supply of VCs relative to the supply of
entrepreneurial ideas a¤ect optimal portfolio size. We show that an increase in the availability of VCs,
keeping all else constant, leads to a reduction in the optimal portfolio size and an improvement in start-
up success. Similarly, an increase in the supply of entrepreneurial projects in the economy results in an
increase in the optimal portfolio size.
Our work is related to several papers in the recent literature. In a recent paper, Inderst, Mueller, and
Muennich (2006) show that VCs may bene�t from limiting the amount of capital they raise by having
�shallow pockets,�since competition for limited funds provides entrepreneurs with stronger incentives, even
if it allows the VC to extract more surplus. Similarly, in our paper competition between entrepreneurs
for the VC�s resources (i.e., his human capital) allows the VC to extract more surplus from his start-ups.
However, in our paper, in contrast to Inderst, Mueller and Muennich (2006), competition for the VC�s
human capital and the VC�s ability to extract more rents a¤ects entrepreneurs�incentives negatively. By
holding a small portfolio, the VC limits the extent of competition between start-ups and commits to extract
lower rents from entrepreneurs, with a positive impact on entrepreneurial incentives. Most importantly,
the main objective of our paper is to investigate the size and focus of a VC�s portfolio. Inderst, Mueller,
and Muennich (2006) abstract from determining the optimal size of the VC�s portfolio by assuming a �xed
number of start-ups and do not consider the bene�ts and costs of having a focused portfolio, a key novel
feature of our model.
Our paper is also related to the work by Kanniainen and Keuschnigg (2003), further extended by
Bernile and Lyandres (2003). In these papers a VC has limited resources that he can devote to his
start-ups, and adding an additional start-up to the portfolio always weakens both the VC�s and the
entrepreneurs�incentives. In our model, adding a new start-up induces competition among start-ups and
allows the VC to extract more surplus at the bargaining stage. However, in our paper, despite the VC�s
higher rent extraction ability, a large portfolio may result in stronger incentives and be bene�cial for both
the entrepreneurs and the VC. This result arises due to the complementarity between entrepreneurial e¤ort
and VC�s investment incentives. In addition, in our paper, the VC�s ability to extract surplus depends
on the degree of relatedness of the start-ups, and thus portfolio focus. Di¤erently from Kanniainen and
Keuschnigg (2003) and Bernile and Lyandres (2003) we derive the optimal size of a VC�s portfolio by
analyzing the combined impact of portfolio size and focus on incentives.
Our work also contributes to the literature stressing the active role of VCs in adding value to their
start-ups, such as Casamatta (2003), Michelacci and Suarez (2004), and Repullo and Suarez (2004), among
4
others. The main di¤erence of our paper from this literature is that these papers consider the incentive
problems between a single VC and a single entrepreneur while in our paper we analyze the VC�s optimal
investment strategy at a portfolio level.
The paper is organized as follows. In Section 2, we describe our basic model. In Section 3.1, we examine
the case where the VC has only one start-up. In Section 3.2, we study the case where the VC has two
start-ups. In Section 3.3, we determine the optimal portfolio size and derive the comparative static results
of the model. In Section 4, we discuss the case in which the VC engages in active portfolio management
by divesting one of his start-ups early. Section 5 presents several extensions of our model and discuss the
robustness of our results. Section 6 provides the empirical implications of our model. Section 7 concludes.
All proofs are in the Appendix.
2 The model
We consider an economy endowed with two types of risk neutral agents: venture capitalists (VCs) and
wealth-constrained entrepreneurs. Entrepreneurs are endowed with a project idea which can be turned,
with the collaboration of a VC, into a �nal marketable product. VCs provide capital as well as other
value adding activities for turning entrepreneurs� ideas into viable businesses. We assume initially that
the VC human capital is a scarce resource in the economy, and that VCs have access to a large supply of
entrepreneurs with project ideas. This assumption re�ects the notion that it takes time and experience to
accumulate skills and human capital to become a VC.4 In section 5, we relax this assumption and study
the impact of VC competition for entrepreneurial start-ups on optimal portfolio size.
Entrepreneurs�project ideas can be turned into a �nal product in two stages. The outcome of the
�rst stage is either a success or a failure. If the �rst stage is successful, then the project is developed and
commercialized during its second stage. If it is a failure, it has no value and is abandoned.
There are four dates in our economy, with no discounting between the dates. At t = 0, the VC chooses
the number � of start-ups to invest in his portfolio. He may invest in either one or two start-ups, or he may
decide to make no investment; thus, � 2 f0; 1; 2g. The development of each start-up requires the active
involvement of both the VC and the entrepreneur. At t = 1; the VC makes a non-contractible start-up
speci�c investment at a personal �xed cost of c, with c > 0. The VC�s investment can be interpreted as the
4For example, on 27 November 2004, The Economist wrote:�perhaps there are simply just a few people in private equity
who are very much better at it than their rivals� in explaining the substantial performance gap across private equity funds.
5
e¤ort of acquiring all the project-speci�c skills and human capital that add value to the start-up, including,
for example, learning about the start-up�s technology and its business opportunities, and developing all
the skills useful in managing the start-up.5 For short, we will refer to these e¤orts as the VC�s start-up
speci�c investment. We assume that the VC�s initial human capital investment with one or two start-ups
has the same cost c: This assumption captures the notion that the VC has only limited time and resources
at his disposal, and that he cannot expand his investment proportionally when he has two start-ups in
his portfolio rather than only one. Thus, given limited resources, the VC can either concentrate all his
resources and human capital on only one start-up, or spread his resources and human capital over two
start-ups, incurring the same cost c in each case.
The VC�s start-up speci�c investment increases the value of the project, and each start-up will have a
higher value with the VC�s investment than without it. For simplicity, we normalize the start-up payo¤
to zero if the VC does not make the initial start-up speci�c investment. Note that this is not a critical
assumption. All we need for our results to hold is that the potential payo¤ from a given start-up is higher
with the VC�s investment than without it. Thus, the VC�s investment and entrepreneurial e¤ort are both
necessary and complementary inputs for the success of each start-up.
Entrepreneurs play a key role during both the �rst and the second stage of their project. At t = 1;
after observing the number of start-ups the VC invests in his portfolio, each entrepreneur exerts e¤ort p;
at a cost of k2p2.6 The parameter k measures the cost of exerting e¤ort, with k > 1: Entrepreneurial e¤ort
determines the success probability of the �rst stage of the project, which becomes known at t = 2.
If the �rst stage of a given project is a failure, the start-up is terminated and both the VC and
the entrepreneur obtain zero payo¤s. If the �rst stage is a success, the second stage needs the active
participation of both the VC and the entrepreneur. If either the entrepreneur or the VC does not participate
to the second stage, the project is divested. For simplicity, we normalize the project�s payo¤ to zero when
divested. We relax this assumption in Section 5, where we allow the entrepreneur to switch to a new VC
and to continue the start-up without the original VC.
5Thus, in our setting, the start-up speci�c investment represents all the non-contractible VC activities that add value to a
venture. For notational simplicity, we do not explicitly consider the VC�s monetary investment into the start-ups. However,
our analysis could easily be modi�ed to incorporate explicitly an initial contractible monetary outlay for each project.6Note also that, while entrepreneurial e¤ort p is modelled as a continuous variable, with p 2 [0; 1], the VC�s input is a
binary choice between making the initial investment, and thus paying the cost c, or not. We make these assumptions for
simplicity, since modelling the VC�s investment as a continuous variable as well considerably reduces the analytical tractability
of the model.
6
We assume that contracts are incomplete in that it is not possible to contract ex-ante on the partici-
pation of either the entrepreneur or the VC to the second stage of the project. This assumption implies
that both the VC and the entrepreneur can withdraw at will their involvement and human capital from
the project at the second stage. Note that this assumption is plausible particularly in the context of VC
investment. Neither the VC nor the entrepreneur(s) can commit ex-ante to the continuation of the project
during its second stage. VCs very often �nance entrepreneurial projects surrounded by great uncertainty.
Not only it is very di¢ cult to describe the �nal outcome of the project ex-ante, but also it is very often
impossible to contract ex-ante on the level of VC�s and entrepreneurs�involvement, the amount of human
capital and resources to be allocated to the project, and the contingencies (such as the state and progress
of the project) under which resources will be available to the start-ups in the future.7
Entrepreneurs�and the VC�s ability to withdraw their human capital from the project implies that it
is not possible to contract on the division of the total surplus between the VC and the entrepreneurs at
t = 0; and that surplus allocation is determined at the interim stage, at t = 2, through bargaining. It
also implies that contracts written ex-ante between the VC and the entrepreneur(s) on how to share the
�nal surplus, such as equity contracts (or options, as in Noldeke and Schmidt, 1998), are ine¤ective since
both the entrepreneur and the VC can (unilaterally) withdraw their participation and human capital from
the implementation phase of the project.8 Even if the VC and the entrepreneur wrote, at the outset of
the venture, a sharing rule on the �nal payo¤ of the start-up, the inability to contract ex-ante on the
participation of the entrepreneur and the VC to the second stage of the project implies that any pre-
existing sharing rule can be renegotiated away, and the division of the surplus is determined entirely by
interim bargaining.
Thus, conditional on observing a successful outcome for the �rst stage of the project at t = 2; the
VC and the entrepreneur(s) bargain over their compensation for the continuation of the start-up. For
simplicity, we assume that the VC and the entrepreneur(s) have equal bargaining power.9 The outcome
of the bargaining process determines the allocation of the surplus between the VC and the entrepreneur,
7Thus, contracts are incomplete in the sense of Hart and Moore (1990) and Grossman and Hart (1986). We recognize that
contracts are a very important aspect of venture capital �nancing in real life. Kaplan and Stromberg (2003) document that
VCs indeed use complex contracts designed to mitigate adverse selection, moral hazard, and hold-up problems. The main
assumption of our paper is that, after all these contractual features are accounted for, VC contracts contain a signi�cant
degree of residual incompleteness and therefore are subject to renegotiation.8For further discussion of this point, see Stole and Zwiebel (1996a) and (1996b).9The more general case where the VC and the entrepreneur have di¤erent bargaining power is available at request from
the authors.
7
and thus a¤ects their incentives.
At t = 3, the payo¤ from the project is realized and distributed between the VC and the entrepreneur(s).
The payo¤ depends on the number of start-ups in the VC�s portfolio. If the VC invests in only one start-up
and concentrates all his initial investment on one start-up only, the payo¤ from the start-up, if successful
in the �rst stage and continued during its second stage, is 2�: If the VC invests in two start-ups and
allocates his initial investment between the two start-ups, each start-up, if successful in the �rst stage and
continued into its second stage, generates a payo¤ of �. Note that this feature is an implication of our
earlier assumption that, if the VC chooses one start-up, he can specialize all his initial investment on one
start-up only, and as a result, the start-up will have a higher payo¤ than in the case if the VC allocates
his initial investment over two start-ups. Since the cost of initial investment, c, is the same whether
the VC invests in one or two start-ups, this assumption implies that the VC can obtain the same total
potential payo¤ from his portfolio, which is then divided among the number of start-ups in the portfolio.
This �linear�payo¤ structure implies that none of our results are driven by the presence of economies or
diseconomies of scale in the VC�s �production technology.�
If one of the start-ups fails in its �rst stage, the VC can concentrate all his resources and human capital
exclusively on the successful start-up, obtaining a payo¤ equal to (1 + �)�; with 0 � � � 1. The value of
the parameter � depends on the ability of the VC to transfer the start-up speci�c investment that he has
made from one start-up to the other. Thus, � depends on the degree of relatedness of the two start-ups,
and we interpret it as representing the degree of �focus�, or scope of the VC�s portfolio.
3 Analysis
3.1 The VC invests in one start-up
Proceeding backward, we �rst characterize the surplus allocation between the VC and the entrepreneur
through interim bargaining. When the VC invests in only one start-up, � = 1, we model the bargaining
game between the entrepreneur and the VC as a standard alternating-o¤ers game where, if bargaining
breaks down, both the VC and the entrepreneur receive their outside options, which we normalize to
zero.10 With equal bargaining power, the VC and the entrepreneur share equally the surplus, 2�, that
they jointly generate and each obtains a payo¤�. Thus, the entrepreneur determines her level of e¤ort p
10See, for example, Binmore, Rubinstein and Wolinski (1986).
8
by maximizing her expected pro�t �1E , that is
maxp
�1E � p�� k
2p2: (1)
Correspondingly, the VC�s expected pro�t �1V is
�1V � p�� c: (2)
Proposition 1 If the VC makes the start-up speci�c investment at t = 1, the optimal level of e¤ort exerted
by the entrepreneur, p1�, is
p1� � �
k: (3)
The corresponding level of expected pro�ts for the VC and the entrepreneur are
�1�V =�2
k� c; �1�E =
�2
2k: (4)
The VC has an incentive to make the start-up speci�c investment only if he expects a positive expected
pro�t. The following lemma characterizes the VC�s investment decision.
Lemma 1 The VC makes the start-up speci�c investment if and only if � � �m �pck:
If the VC does not make the investment at t = 1 (that is, if � < �m), the payo¤ from the project is
zero, the entrepreneur does not exert any e¤ort, and both parties obtain zero pro�ts.
3.2 The VC invests in two start-ups
We proceed again backward, and �rst characterize the outcome of the interim bargaining game between
the VC and the two entrepreneurs. When the VC invests in two start-ups, � = 2, the bargaining process
between the VC and the entrepreneurs depends on whether only one or both projects have a successful
outcome at their �rst stage, t = 2: There are three di¤erent possible cases (states of the world): (i) both
projects are successful in their �rst stage, state SS; (ii) one project is successful while the other one is a
failure, state SF;11 (iii) both projects are a failure, state FF .
We begin our analysis with the (simpler) case where only one start-up, say start-up i, is successful,
state SF . In this case, the VC can reallocate his human capital and concentrate exclusively on start-up i,
increasing its payo¤ from � to (1 + �)�. Since the VC has only one successful start-up in his portfolio,
11Note that, given that the two entrepreneurs are identical, it is irrelevant which one of the two projects is
successful. Thus, we will treat these two separate but symmetric cases e¤ectively as a single case.
9
the entrepreneur and the VC engage in bargaining with alternating o¤ers (with no outside options) as
in the previous section. Thus, the VC�s and the entrepreneurs�payo¤s, denoted respectively by l2V (SF ),
l2Ei(SF ) and l2Ej(SF ), are
l2V (SF ) =(1 + �)�
2; l2Ei(SF ) =
(1 + �)�
2; l2Ej (SF ) = 0: (5)
If both start-ups are successful at t = 2, state SS, the VC bargains with both entrepreneurs. We model
this process of �multilateral�bargaining between the VC and the two entrepreneurs as in Stole and Zwiebel
(1996a). We assume that the VC leads individual bargaining sessions with one start-up at a time, starting,
say, with entrepreneur i. Individual bargaining occurs again as an alternating o¤er game. If the VC and
entrepreneur i reach an agreement, the VC starts a round of bargaining with entrepreneur j. If bargaining
between the VC and entrepreneur i breaks down without an agreement, entrepreneur i drops from the
bargaining process and the VC engages in bargaining with entrepreneur j; where both players have zero
outside options. Equilibrium payo¤s are subject to the (stability) condition that if the VC reaches an
agreement with entrepreneur i, and bargaining with entrepreneur j breaks down, then entrepreneur j
drops from the bargaining process and the VC and entrepreneur i renegotiate their original agreement
through bargaining, where now both players have zero outside options. This condition ensures that the
agreement reached between the VC and entrepreneur i (resp. j) anticipates the renegotiation that would
take place if bargaining between entrepreneur j (resp. i) and the VC breaks down. The VC�s and the
entrepreneurs� payo¤s, denoted respectively by l2V (SS) � l2Vi(SS) + l2Vj (SS), l2Ei(SS) � � � l2Vi(SS);
i = 1; 2, are characterized by (see Stole and Zwiebel, 1996a)
�� l2Vi(SS) = l2V (SS)� �ljV ;
�� l2Vj (SS) = l2V (SS)� �liV ;
where
�liV �(1 + �)�
2(6)
represents the VC�s outside option when bargaining with entrepreneur j. Solving the above system of
equations, we obtain
l2V (SS) =(3 + �) 2�
6; l2Ei(SS) =
(3� �)�6
; i = 1; 2: (7)
By examining the VC�s payo¤ in the SS state (7), it is easy to see that when the VC has two successful
start-ups in his portfolio he obtains a greater fraction of the total surplus than when he has only one
10
successful start-up, that is, (3+�)6 � 12 for all 0 � � � 1.12 This happens because having a second successful
start-up in his portfolio gives the VC an outside option while bargaining with each entrepreneur. Hence,
the presence of a second start-up and the ability to transfer resources from one start-up to the other
creates �competition�between entrepreneurs, allowing the VC to extract more surplus. The VC�s ability
to transfer ex-post resources from one start-up to another is critical, since when � = 0 the VC extracts the
same fraction of the surplus with both one and two start-ups. Note also that the VC�s surplus, l2V (SS),
is increasing in the degree portfolio focus, � whereas each entrepreneur�s surplus, l2Ei(SS), is decreasing
in the level of portfolio focus, �. Thus, a greater level of portfolio focus � bene�ts the VC but hurts the
entrepreneurs at the bargaining stage. This happens because a greater level of focus leads to a greater
outside option for the VC while he bargains with each entrepreneur, allowing the VC to extract a greater
fraction of the total surplus. Note however that, as we will show below, the entrepreneurs will bene�t
ex-ante from a greater degree of focus.
Finally, if both entrepreneurs fail in the �rst stage, state FF , both start-ups are terminated and all
agents obtain zero payo¤s.
We now characterize the entrepreneurs�choice of e¤ort. If the VC makes the speci�c investment for
each start-up, anticipating her payo¤s in di¤erent states of the world, entrepreneur i determines her e¤ort
level by maximizing her expected pro�t �2Ei , that is
maxpi
�2Ei � pipj(3� �)�
6+ pi(1� pj)
(1 + �)�
2� k
2p2i ; i; j = 1; 2; i 6= j: (8)
Similarly, the VC�s expected pro�t �2V is
�2V � pipj(3 + �) 2�
6+ pi(1� pj)
(1 + �)�
2+ pj(1� pi)
(1 + �)�
2� c; i; j = 1; 2; i 6= j: (9)
The �rst-order condition of (8) is
pi(pj) =(3 (1 + �)� 4�pj)�
6k: (10)
Note that the e¤ort exerted by entrepreneur i decreases in the e¤ort exerted by entrepreneur j and hence,
the e¤ort levels are strategic substitutes. This happens because, when the VC has two start-ups in his
portfolio, in state SS he extracts a higher surplus from each entrepreneur, reducing their expected pro�ts
and their incentives to exert e¤ort.12Note also that the payo¤s in (7) are the same as the players�Shapley value of the corresponding cooperative game.
11
Proposition 2 If the VC makes start-up speci�c investment for each start-up at t = 1, the Nash-
equilibrium level of e¤ort, denoted by p2�, is
p2� � 3(1 + �)�
2 (2��+ 3k): (11)
The corresponding level of expected pro�ts for the VC and the entrepreneurs are
�2�V =3(��+ 3k) (1 + �)
2�2
2 (2��+ 3k)2 � c; (12)
�2�E1 = �2�E2 =
�3((1 + �)�)
2(2��+ 3k)
�2k
2: (13)
It is easy to verify that the equilibrium level of entrepreneurial e¤ort p2�, is increasing in the degree
of portfolio focus, �. The focus parameter � has two opposing e¤ects on the level of e¤ort chosen by
the entrepreneur. On the one hand, a higher degree of focus allows the VC to extract more surplus from
each entrepreneur in the SS state, with a negative e¤ect on entrepreneurial e¤ort. On the other hand, a
more focused portfolio allows the VC to reallocate more e¢ ciently his resources to the successful start-up
in the SF state, where only one of the start-ups is successful in its �rst stage, with a positive e¤ect on
entrepreneurial e¤ort. As it turns out, the second e¤ect dominates the �rst e¤ect and the overall impact
of an increase in focus on the level of e¤ort and the expected pro�ts is always positive.
Entrepreneurial incentives to exert e¤ort and, in turn, the VC�s investment incentives depend on the
number of start-ups in the VC�s portfolio. An important question is whether the entrepreneurs have
stronger incentives to exert e¤ort when the VC has one or two start-ups in his portfolio.
Lemma 2 If the VC is induced to make the necessary start-up speci�c investment with both one and two
start-ups in his portfolio, each entrepreneur always has greater incentives to exert e¤ort when his start-up
is the only start-up in the VC�s portfolio:
p1� > p2�:
Furthermore, the di¤erence between the levels of e¤ort, p1� � p2�, increases in project payo¤, �, and
decreases in the degree of focus, �, and in the entrepreneur�s cost of exerting e¤ort, k.
If the VC makes the start-up speci�c investment, entrepreneurial incentives to exert e¤ort are always
lower when the VC has two start-ups rather than when he has only one. This is due to the fact that,
with two start-ups in his portfolio, the VC adds less value to each start-up and is able to extract more
surplus from the entrepreneurs. Thus, conditional on the VC making the start-up speci�c investments,
12
entrepreneurial incentives are always worse when the start-ups belong to a large portfolio, leading to a
lower level of e¤ort.13
The di¤erence between the level of e¤ort in the two cases is increasing in the project payo¤, �. This
property is due to the fact that when the VC has two start-ups each entrepreneur bene�ts less from an
increase in the project payo¤ since the VC can extract a greater fraction of the incremental surplus from
the entrepreneurs. The di¤erence between the level of e¤ort in a small and large portfolio is decreasing in
the level of focus, �. This can be seen by noting that an increase in the degree of focus, �, increases p2�
while it has no e¤ect on p1�, reducing the di¤erence between the two e¤ort levels. Finally, an increase in
the cost of e¤ort, k, always reduces entrepreneurial e¤ort, but relatively more when the VC has only one
start-up.
It is important to note that, due to the complementarity between entrepreneurial e¤ort and the VC�s
investment, it is possible that each entrepreneur exerts greater e¤ort when the VC has a large portfolio
rather than a small one, despite the fact that they obtain lower rents in a larger portfolio. This is because,
under some conditions, the VC has incentives to make the start-up speci�c investment only when he holds
a large portfolio, leading the entrepreneurs to exert e¤ort as well. More speci�cally, this possibility arises
when the VC�s expected pro�t from investing in a single start-up is negative, while his expected pro�t
from investing in two start-ups is positive. In this case, the VC can recover his initial cost c only if he
holds a large portfolio. Hence, the VC will be willing to incur the cost of his initial investment and the
entrepreneurs will exert e¤ort only if the VC invests in two start-ups. We will elaborate on this possibility
in more detail in the next section.
3.3 Optimal portfolio size
The VC chooses his portfolio size as a result of the interaction of three distinct e¤ects and their impact on
incentives. The �rst one is the rent extraction e¤ect: the VC can extract greater rents when he has a larger
portfolio. This e¤ect always induces the VC to prefer (all else equal) a larger portfolio. The second e¤ect
is the resource allocation e¤ect: by investing in two, rather than only one start-up, the VC can reallocate
ex-post his resources and human capital from one start-up to the other. The strength of this e¤ect depends
on the degree of focus of the portfolio, �. If the success probability of each start-up is �xed and the same
13Note that each entrepreneur has an incentive to exert e¤ort only if she expects the VC to make the start-up speci�c
investment. In turn, the VC is willing to make the start-up speci�c investments only if he expects a positive pro�t, net of
his total investment cost c:
13
regardless of whether the VC has one or two start-ups, this e¤ect always leads the VC to prefer a large
portfolio to a small one.14 The third e¤ect is the value dilution e¤ect: a larger portfolio requires the VC
to spread his �xed amount of resources and human capital over a larger number of start-ups. The result
is that the VC adds lower value to each start-up: he will add only � to each start-up in the SS state, and
(1 + �)� in the SF state. Thus, the value dilution e¤ect favors a small portfolio.
Portfolio size a¤ects the VC�s and the entrepreneurs�incentives as follows. First, in a large portfolio,
the rent extraction e¤ect favors the VC and, thus, impacts entrepreneurial incentives negatively and the
VC�s investment incentives positively. Second, a large portfolio allows the VC to reallocate his human
capital from one start-up to another in case one of the start-ups fails; this possibility bene�ts both the
entrepreneurs and the VC and, thus, a¤ects their incentives positively. Furthermore, this e¤ect is stronger
when the level of the portfolio focus is higher. Third, in a large portfolio, dilution from spreading the VC�s
resources and human capital over two start-ups lowers the payo¤ from exerting e¤ort, and thus reduces
both entrepreneurial e¤ort and the VC�s investment incentives.
The VC�s optimal portfolio size depends on the value of the project payo¤, �, and portfolio focus,
�, which may fall in one of three possible regions (see Figure 1): the VC can invest in no start-up at all
(Region 0), in one start-up (Region 1), or in two start-ups (Region 2), as summarized in the following
proposition.
Proposition 3 There are critical values f�c;�1(�; k);�2(�; k)g (de�ned in the appendix) such that the
VC�s optimal portfolio size is as follows:
i) for low project payo¤ (0 � � < �1) the VC invests in no start-up, �� = 0 (Region 0);
ii) for high project payo¤ (� � �2) the VC invests in one start-up only, �� = 1 (Region 1);
iii) for moderate project payo¤ and high focus (�1 � � < �2 and �c � � � 1) the VC invests in two
start-ups, �� = 2 (Region 2).
Furthermore, @�1(�;k)@� � 0 , @�2(�;k)
@� � 0, and @�2(�;k)@k � 0.
14This property can be seen as follows. If the VC has only one start-up, and the entrepreneur exerts e¤ort p, total
expected value of the VC�s portfolio is p2�: If the VC has two start-ups in his portfolio, and each entrepreneur
exerts e¤ort q, the total expected value of the VC�s portfolio is 2q2�+2q(1� q)(1+�)� = 2q�+2q(1� q)��. Itis easy to see that the expected value of the VC�s portfolio is larger with two start-ups than with only one when
� > p�qq(1�q) ; which is always the case when p = q and � > 0. Note, however, that in our analysis, p and q are
determined endogenously as a function of portfolio size and focus.
14
Two key insights emerge from Proposition 3. The �rst is that the VC �nds it optimal to hold a small
portfolio when the project payo¤, �, is relatively high, that is, when � � �2 (Region 1). In this region,
the bene�ts of a small portfolio in terms of better entrepreneurial incentives dominate the advantages of
a large portfolio in terms of rent extraction and resource reallocation. The intuition for this result is as
follows. From Lemma 2, we know that the di¤erence in the entrepreneurs� e¤ort levels in a small and
a large portfolio, that is, p1� � p2�, is greater when project payo¤ � is larger. This implies that the
negative impact on entrepreneurial incentives of holding a large portfolio is greater at higher values of
�. The proposition shows that, for start-ups with a large �, the VC prefers to give up the greater rent
extraction ability and the resource allocation advantages of a large portfolio for the bene�ts of stronger
entrepreneurial incentives of a small portfolio.
Note that the parameter � can be interpreted as representing the project�s residual expected value,
after the �rst stage is completed. Thus, a greater value of � characterizes start-ups with either larger
ultimate payo¤, or with greater ultimate success probability. In other words, greater values of � represent
start-ups with stronger ex-ante fundamentals. Proposition 3, therefore, implies that the VC �nds it optimal
to have a smaller portfolio when he has access to start-ups with strong fundamentals. By holding a small
portfolio, the VC boosts entrepreneurial incentives and increases the success probability of his start-ups.
A small size portfolio, therefore, is desirable precisely because it allows the VC to obtain superior ex-post
performance from his investment.
The second insight of Proposition 3 is that the VC �nds it optimal to hold a large portfolio when
the project payo¤ � is moderate and when he can form a portfolio with su¢ cient focus, that is, when
�1 � � < �2 and �c � � � 1 (Region 2). For start-ups with a more moderate potential, the bene�ts of a
larger portfolio in terms of greater rent extraction and resource reallocation ability dominate the incentive
advantage of a small portfolio. The intuition for this result is as follows. In this region, a large value
of � implies that the rent extraction and resource reallocation e¤ects of a large portfolio are signi�cant.
Furthermore the di¤erence between the levels of entrepreneurial e¤ort in a small and a large portfolio is
smaller at moderate values of �, as established in Lemma 2. Thus, when parameter values fall in this
region, the rent extraction and resource allocation e¤ects dominate the incentive e¤ect, and the VC holds
a large portfolio. This also implies that, for a given project payo¤�, the VC �nds it optimal to expand his
portfolio only at su¢ ciently high values of � (see again Figure 1). Thus, the VC is willing to increase the
size of his portfolio only if he can form a portfolio with su¢ cient focus by investing in highly related start-
ups, a property which is formally re�ected by the fact that @�2(�;k)@� � 0. In addition, @�2(�;k)
@k � 0 implies
15
that an increase in the entrepreneurs�cost of exerting e¤ort, k, makes larger portfolios more desirable.
This happens because an increase in the value of k reduces entrepreneurial e¤ort, which leads to a lower
success probability for each start-up and to a riskier portfolio. As a result, the VC�s willingness to hold
a larger portfolio increases since a lower success rate for each start-up increases the importance of the
resource reallocation bene�t of large portfolios. This also implies VCs prefer larger and more focused
portfolios for start-ups with moderate fundamentals.
Note also that for some parameter values in this region, that is, when�1 � � < �m, the VC�s expected
pro�ts are positive only if he holds a large portfolio, and negative if he holds a small portfolio (that is,
�1�V < 0 and �2�V > 0). This happens because, with a small portfolio, the VC cannot extract enough rents
from the entrepreneur to compensate him for the cost of making the initial investment, c. In this case,
anticipating that the VC is not willing to make the initial start-up investment, the entrepreneur does not
exert e¤ort either, and the project is not undertaken even if it is potentially pro�table. Investing in a
large portfolio, however, provides the VC with the rent extraction and resource reallocation bene�ts and
induces him to make the required initial investment. Anticipating the improved incentives of the VC, the
entrepreneurs exert e¤ort and the projects become viable. As a result, both the VC and the entrepreneurs
turn out to be better o¤ when the VC holds a large portfolio.
The above result has an interesting implication that entrepreneurial ideas with a moderate value may
be economically viable only if the VC can form a large portfolio with a su¢ cient degree of focus. If we
interpret � as measuring the size of a start-up, this implies that VCs would be willing to invest in small
businesses only if they are able to combine such start-ups in a portfolio of su¢ cient size and focus. It also
implies that entrepreneurs with smaller businesses will have an incentive to cluster in similar or related
industries so that they can be �nanced by a common VC. Thus, small and risky businesses (characterized by
small or moderate �) may be economically viable and obtain VC �nancing only if they have a su¢ cient
degree of industry focus among them. A policy implication from this result is that encouraging small
business creation in the same or related industries will improve available VC �nancing and enhance social
welfare because potential VCs will be willing to provide monetary and human capital to such businesses
only if they have a common industry focus.
Finally, when the level of � is very low, that is when 0 � � < �1 (Region 0), the VC does not invest in
any start-up. In this region the project payo¤ is so low that the VC cannot recover his initial investment
cost c. As a result, the VC does not make any start-up speci�c investment and the entrepreneurs do not
exert any e¤ort. Hence the project opportunities cannot be exploited.
16
4 Portfolio Management
In this section we show that the VC can increase his expected pro�ts by engaging in active portfolio
management, that is, by divesting one of his successful start-ups. This strategy can be optimal since the
possibility of divesting one of the start-ups allows the VC to extract more surplus from the remaining one.
Thus, this section helps shed some light on why VCs may make seemingly socially ine¢ cient decisions by
terminating some of their start-ups prematurely in order to maximize their own welfare.15
We modify our basic model as follows. Consider the case in which the VC invests in two start-ups and
both entrepreneurs have a successful �rst stage, state SS. The VC now faces two choices. He can either
continue both start-ups, or divest one of them and dedicate himself entirely to the remaining one. If the
VC chooses not to continue a successful start-up, he can divest it, for example through a sale to another
VC (or some private buyer), or even take it public in an IPO. We assume that the proceeds from divesting
the start-up are lower than the proceeds from continuing with the original VC and, for simplicity, are
normalized to zero.16 The VC now has the option to bargain with one entrepreneur for the continuation
of only her start-up and the termination of the other start-up, which we denote as �bilateral�bargaining.
Alternatively, the VC can engage, as before, in multilateral bargaining with both entrepreneurs for the
continuation of both start-ups. This choice is important because the VC may be able to extract a di¤erent
surplus depending on whether he continues one or both start-ups.
We model the process of bilateral bargaining in state SS as follows. The VC selects, with equal
probability, one of the two successful start-ups, say start-up i, and negotiates with entrepreneur i the
payo¤ that he will receive for his exclusive participation to the continuation of start-up i only. This may
be achieved, for example, by negotiating, at the bargaining stage, an agreement between the VC and the
entrepreneur that limits the VC�s ability to participate in other start-ups. This implies that the VC can
commit not to participate to the continuation of the other project and to divest it. The VC�s ability to
make such a commitment at this stage of the game (after the realization of the state of the world) is a
much weaker requirement than the assumption that the VC can, at the beginning of the game, commit
to continue a start-up under predetermined circumstances, a possibility that we have ruled out.17 While
15See, for example, the Economist, November 27, 2004 which reports that �Google�s founders would have preferred to wait
longer to do their IPO, but had to rush it because venture capitalists, including Kleiner Perkins, wanted to cash in.�16Normalizing divestiture payo¤ is only a simpli�cation. Our results go through as long as divestiture payo¤ is lower than
the payo¤ possible with the incumbent VC. This assumption re�ects that the incumbent VC, because of the initial speci�c
investment he made, can generate a greater payo¤ than the divestiture payo¤.17Note that the use of such provisions is very common in stock purchase agreements. For a discussion of covenants in
17
bargaining with entrepreneur i, the VC has the outside option to go to the start-up j and start a new
round of bargaining with entrepreneur j, obtaining �ljV , de�ned in (6). Both entrepreneurs have again zero
outside options. This implies that the VC�s and entrepreneur i�s payo¤s are given by
l2BV (SS) � �ljV +1
2
h(1 + �)�� �ljV
i=3 (1 + �)�
4; (14)
l2BEi (SS) � (1 + �)�� l2BV (SS) =(1 + �)�
4; l2BEj (SS) = 0: (15)
A critical question is whether the VC can obtain a greater payo¤ by continuing both start-ups or by
divesting one of them and continuing only the remaining one. This choice depends on whether the VC
can extract more surplus by engaging in multilateral or bilateral bargaining with the entrepreneurs, given
the structure of the bargaining games.18
Proposition 4 The VC continues only one project and divests the other if and only if � � 35 .
When the VC divests one of the start-ups, from (14), he will receive 34 of the total surplus (1 + �)�.
When the VC continues both start-ups, from (7), he will receive a fraction 3+�6 of the total surplus 2�. By
direct comparison, it is easy to see that 34 >3+�6 for all 0 � � � 1. This implies that, divesting one of the
start-ups increases the VC�s rent extraction ability, but at the cost of reducing total surplus from 2� to
(1 + �)�. Proposition 4 states that when the loss from divesting one of the start-ups is not too large, that
is, when � � 35 , the VC �nds it optimal to exploit the better bargaining position provided by bilateral
bargaining, and thus prefers to continue one start-up only. When there is no value loss in reallocating
resources from one start-up to the other, that is, when � = 1, the VC always prefers to divest one of the
two start-ups. In contrast, when the loss from divesting one project is su¢ ciently large, � < 35 , the VC
prefers to give up the better bargaining position provided by bilateral bargaining in order to realize the
full potential of his portfolio and continues both start-ups.
If � < 35 , the VC will choose in the SS state to continue both start-ups, and Proposition 3 will remain
valid. Thus, in the remainder of the analysis, we will assume that � � 35 and characterize the optimal
portfolio size when the VC divests one of the successful start-ups in the SS state. In the SF and FF
states, the game unfolds as before.
shareholders agreements, see Gompers and Lerner (1996) and Chemla, Habib and Ljungqvist (2006).18 In other words, given the structure of the respective bargaining games, the VC may be willing to ine¢ ciently terminate
one of the two start-ups if he cannot internalize a su¢ ciently large part of the e¢ ciency gains that can be obtained by
continuing both start-ups. The question of whether the VC can internalize, through multilateral bargaining, a su¢ cient
portion of the e¢ ciency gains to induce him to always make the socially e¢ cient decision is ultimately an empirical one.
18
Anticipating her payo¤s in di¤erent states of the world, entrepreneur i determines her level of e¤ort pi
by maximizing her expected pro�t �2BEi , that is
maxpi
�2BEi � pipj1
2
(1 + �)�
4+ pi(1� pj)
(1 + �)�
2� k
2p2i ; i; j = 1; 2; i 6= j: (16)
Correspondingly, the VC�s expected pro�t �2BV is
�2BV � pipj3(1 + �)�
4+ pi(1� pj)
(1 + �)�
2+ pj(1� pi)
(1 + �)�
2� c; i; j = 1; 2; i 6= j: (17)
Proposition 5 The Nash-equilibrium level of e¤ort, denoted by p2B�; is
p2B� � 4(1 + �)�
3(1 + �)� + 8k: (18)
The corresponding level of expected pro�ts for the VC and the entrepreneurs are
�2B�V =8((1 + �)� + 4k) (1 + �)
2�2
(3(1 + �)� + 8k)2� c; (19)
�2B�E1 = �2B�E2 =8k (1 + �)
2�2
(3(1 + �)� + 8k)2: (20)
The following proposition characterizes the VC�s optimal portfolio size.
Proposition 6 Let 35 � � � 1 and c < �c(k) (de�ned in the appendix). There are critical values
f�B1 (�; k);�B2 (�; k)g (de�ned in the appendix) such that the VC�s optimal portfolio size is as follows:
i) for low project payo¤ (0 � � < �B1 ) the VC does not invest in any start-up, �� = 0;
ii) for high project payo¤ (� � �B2 ) the VC invests in one start-up only, �� = 1;
iii) for moderate project payo¤ and high focus (�B1 � � < �B2 and 35 � � � 1) the VC invests in two
start-ups, �� = 2.
Furthermore, @�B1 (�;k)@� � 0 and @�B
2 (�;k)@� > 0:
Proposition 6 mirrors Proposition 3, with one di¤erence that when both start-ups are successful (state
SS) the VC will divest one of them. The following proposition compares Proposition 3 and Proposition 6
and illustrates the e¤ect of portfolio management on the VC�s portfolio size.
Proposition 7 Let 35 � � � 1 and c < �c(1; k). There is a �̂ such that:
i) if �̂ < � � 1 : �B1 < �1 and �B2 > �2;
ii) if3
5� � � �̂ : �B1 � �1 and �B2 � �2:
19
When start-ups are highly related, that is, when �̂ < � � 1, the region in which the VC invests in
two start-ups is larger with portfolio management, implied by �B2 > �2. This happens because, when
� is large the VC can form a focused portfolio and can divest one of the successful start-ups with little
e¢ ciency loss. Thus, active portfolio management, by enhancing the VC�s ex-post rent extraction ability,
induces him to choose a larger portfolio ex-ante. Note that the greater rent extraction ability due to active
portfolio management also means that the VC has stronger incentives to make the initial start-up speci�c
investment at t = 1. Thus, active portfolio management enlarges the set of feasible start-ups, implied
by �B1 < �1. This means that there are start-ups with a low payo¤, i.e., those with � 2 [�B1 ;�1),
that are economically viable only when the VC can increase his rent extraction ability by engaging in
portfolio management.19 Hence, even though portfolio management results ex-post in a socially ine¢ cient
termination of a start-up, it may nevertheless improve ex-ante social welfare by enlarging the set of feasible
start-ups.
The above result is reversed when start-ups are only moderately related, that is, when 35 � � � �̂. In
this case portfolio management shrinks the region where the VC invests in two start-ups, implied by �B2 �
�2. Divesting a successful start-up is now more costly, because the VC cannot reallocate his resources
e¢ ciently from one start-up to the other. Since the VC anticipates the cost of ine¢ cient termination
of start-ups, he chooses a smaller portfolio ex-ante. Correspondingly, for this range of parameters, the
possibility of active portfolio management shrinks the set of feasible start-ups compared to the case where
the VC cannot engage in portfolio management, implied by �B1 � �1.
5 Extensions and robustness
In our basic model we study the static portfolio problem of a single VC facing a large supply of entrepre-
neurial projects. In this section we extend our model in two directions: First, we introduce VC competition
for entrepreneurs and examine its e¤ect on optimal portfolio size. Competition among VCs for start-ups
generates an outside option for entrepreneurs and, thus, a¤ects both entrepreneurial incentives and the
optimal portfolio size. Second, we consider a dynamic, in�nite-horizon version of our basic model, where
the VC can invest in new start-ups over time when a project either fails or is completed. We show that
the main results of our static model carry over to these more general settings. For analytical tractability
19Note that, in this case, the entrepreneurs as well are ex-ante better-o¤. This is because the VC will be willing to �nance
their start-ups only if he can engage in portfolio management, even if this implies that one of the start-ups is divested ex
post in the SS state.
20
we analyze these two cases separately and, for notational simplicity, we set k = 1.
5.1 Competition for entrepreneurs
This section extends our basic model such that VCs compete for entrepreneurs �rst ex-ante, when they
form their portfolios at t = 0, and then a second time in the interim, when the �rst stage of the project
is completed. We modify our model as follows. First, we assume that a successful entrepreneur can, at
t = 2, turn to a new VC and still continue her start-up in the second stage. The ability to complete the
project with an alternative VC gives the entrepreneur an outside option while bargaining with the original
VC. Switching to a new VC, however, results in a less valuable start-up than staying with the original
VC that made the initial start-up speci�c investment. This assumption re�ects the notion that, because
of the relationship-speci�c nature of the human capital investment made at t = 1, the surplus created
from switching to a new VC is strictly smaller than the surplus obtained by continuing with the original
VC. This implies that the entrepreneur and the original VC will �nd it in their best interest to reach an
agreement at the interim bargaining stage despite the entrepreneur�s ability to continue with a new VC.
Formally, we assume that if an entrepreneur switches to a new VC, she obtains a payo¤ of �12� (with
0 < �1 < 1) in the case where the original VC has one start-up, and �1� in the case where the original
VC has two start-ups in his portfolio. The VC has the same outside options as in the basic game.
If at t = 0 the VC invests in only one start-up, at the interim bargaining stage the entrepreneur
obtains ~l1E � (1+ �1)� due to her ability to switch to a new VC and the VC obtains ~l1V � (1� �1)�. The
entrepreneur chooses her level of e¤ort p to maximize her expected pro�t ~�1E , that is
maxp
~�1E � p(1 + �1)��p2
2; (21)
resulting in ~p1� � (1+ �1)�. The corresponding level of expected pro�ts for the VC and the entrepreneur
are
~�1�V = (1� �21)�2 � c; ~�1�E =(1 + �1)
2�2
2: (22)
If at t = 0 the VC invests in two start-ups, he will bargain again with the two entrepreneurs, each having
the option to leave the original VC and switch to another VC. By modifying the multilateral bargaining
game discussed in the basic model correspondingly, it is easy to show that each entrepreneur�s payo¤ from
bargaining in the SS state, denoted by ~l2Ei(SS) and in the SF , denoted by~l2Ei(SF ) state, are
~l2Ei(SS) =�
2+�(�1 � �)
6; ~l2Ei(SF ) =
�
2+�(2�+ �1)
4i = 1; 2;
21
respectively. Thus, entrepreneur i determines her e¤ort level pi to maximize her expected pro�t ~�2Ei , that
is,
maxpi
~�2Ei � pipj(�
2+�(�1 � �)
6) + pi(1� pj)(
�
2+�(2�+ �1)
4)� pi
2
2: (23)
The Nash-equilibrium e¤ort level, ~p2�, is
~p2� � 3�(2(1 + �) + �1)
� (�1 + 8�) + 12: (24)
The corresponding expected pro�ts for the VC and the entrepreneurs are
~�2�V =6�2 (2(1 + �) + �1) (2 (1 + �) (3 + ��) + �1(� (1� 2�)� 3))
(12 + (�1 + 8�)�)2 ;
~�2�E1 = ~�2�E2 =9 (2(1 + �) + �1)
2�2
2 (12 + (�1 + 8�)�)2 :
Note that the ability to switch to a new VC allows the entrepreneurs to extract more surplus from
the original VC, with a positive e¤ect on their e¤ort incentives and, thus, the success probability of the
start-ups, as summarized in the following lemma.
Lemma 3 Entrepreneurial e¤ort is strictly increasing in �1.
In addition to ex-post competition, we assume that VCs compete for entrepreneurs ex-ante when
they choose their portfolios at t = 0. Availability of other VCs is important if entrepreneurs and VCs
bargain ex-ante for the division of the joint surplus. Ex-ante competition for entrepreneurs provides the
entrepreneurs with an outside option which a¤ects the division of the total ex-ante expected surplus.
We further modify the basic game as follows. At the beginning of the game, t = 0, the VC selects
one or two start-ups and bargain with the entrepreneur(s). Bargain results in a side payment, t, from the
VC to the entrepreneur(s). We assume that t � 0 since we continue to assume that the entrepreneurs are
wealth constrained. Entrepreneurs have now an outside option with a value of �0 � 0. We interpret �0 as
a measure of the scarcity of entrepreneurs relative to VCs, and therefore the tightness of the VC market.20
The VC moves �rst and chooses whether to form a small or a large portfolio, and therefore, whether to
bargain with one or two entrepreneurs. The following proposition characterizes the optimal portfolio size
when the entrepreneurs have outside options both ex-ante, �0, and in the interim, �1.
20Note that we assume that at the beginning of the game the VC and the entrepreneurs have already identi�ed each other
and that the value for the entrepreneurs to search for alternative VCs is lower than the expected payo¤ they can obtain from
bargaining with the present VC. This is because the search for alternative VCs is costly.
22
Proposition 8 There are critical values f~�c; ~�1; ~�2g (de�ned in the appendix) such that the VC�s optimal
portfolio size is as follows:
i) for low project payo¤ (0 � � < ~�1) the VC does not invest in any start-up, �� = 0;
ii) for high project payo¤ (� � ~�2) the VC invests in one start-up only, �� = 1;
iii) for moderate project payo¤ and high focus ( ~�1 � � < ~�2 and ~�c � � � 1) the VC invests in two
start-ups, �� = 2.
Furthermore, @~�2
@�0< 0.
Proposition 8 shows that our earlier results hold also in the case when entrepreneurs can switch to
alternative VCs both ex-ante and in the interim. At the outset of the game, the VC chooses the size of
his portfolio by comparing the bene�ts of large and small portfolios identi�ed in the basic model. When
start-ups have a high project payo¤, �, the bene�ts of a small portfolio in terms of better entrepreneurial
incentives outweigh the bene�ts of large portfolio in terms of rent extraction and resource reallocation
ability. Thus, the VC optimally chooses a small portfolio and sets �� = 1. For moderate value of project
payo¤ and high portfolio focus, the VC prefers a large portfolio and sets �� = 2.
The di¤erence with the basic model is that the entrepreneurs�outside option due to the availability
of other VCs a¤ects the ex-ante allocation of the surplus, and the portfolio size. The ex-ante availability
of alternative VCs creates competition for the VC and a¤ects him negatively by reducing his ability to
extract surplus from the entrepreneurs. The magnitude of this negative e¤ect is greater (at the margin) in
a large portfolio than in a small one. This happens because in a large portfolio the VC must bargain with
a greater number of entrepreneurs, each of whom possesses an outside option, �0. Thus, the availability
of other VCs, or an increase in the supply of VCs relative to the supply of entrepreneurs, reduces the
desirability of large portfolios. This means that VCs are more likely to choose a smaller portfolio when
entrepreneurs have a greater outside option, that is, when entrepreneurs are relatively more scarce than
VCs, implied by @ ~�2
@�0< 0. Correspondingly, VCs are more likely to have larger portfolios when there is an
increase in the supply of entrepreneurial projects, re�ected by a reduction in �0:
5.2 A dynamic extension
In this section we discuss a dynamic, in�nite-horizon version of our basic model, where the VC can invest
in new start-ups when the existing projects in his portfolio fail or are completed. At t = 0, the VC can
23
invest in either one or two start-ups, or none: �0 2 f0; 1; 2g. If the VC invests in one start-up, and the
start-up fails at t = 1, the VC can invest in a new start-up. Investment in a new start-up will require
the VC to make a new start-up speci�c investment. For notational simplicity, we assume that the VC
sustains the same cost c every time he invests in new start-ups.21 If the start-up is successful, the VC
and the entrepreneur complete the second stage of the project, as in the static model. The di¤erence from
the static model is that now, while bargaining with the entrepreneur, the VC has the ability to invest in
a new start-up. This ability allows the VC to extract more surplus from the entrepreneur. Letting v1
be the VC�s continuation payo¤ in the (sub)game, at the bargaining stage the VC�s surplus is given by
l̂1V = v1 +(2���v1)
2 . The entrepreneur�s surplus is given by l̂1EN = (2���v1)2 , and she chooses her e¤ort
level p to maximize her expected pro�t �̂1DE , that is
maxp
�̂1DE � p
2(2�� � v1)�
p2
2; (25)
resulting in p̂1� � �� � v12 . Stationarity implies that the VC�s expected pro�t from holding a small
portfolio, v1, is implicitly de�ned by
v1 = �
�p̂1�(v1 +
1
2(2�� � v1) + �v1) + (1� p̂1�)v1
�� c; (26)
where � < 1 is the appropriate discount factor.
Now consider the case where the VC invests in two start-ups at t = 0: If one of the two start-ups
fails, state SF , we assume (for simplicity) that it is always optimal for the VC to reallocate his human
capital to the successful start-up rather than to invest in an additional one. In this case, the VC and
the entrepreneur bargain over the division of the surplus under the condition that the VC has the outside
option of investing new start-ups. This implies that the VC will obtain a surplus l̂2V (SF ) =(1+�)��+v2
2 ,
leaving l̂2E(SF ) =(1+�)���v2
2 to the entrepreneur. Similarly, if both start-ups are successful, state SS,
the VC�s bargaining with both start-ups proceeds as described in the basic model, under the assumption
that if bargaining with one start-up breaks down, the VC reallocates his human capital to the remaining
start-up and starts a fresh round of bargaining. Bargaining with the remaining start-up takes place again
under the condition that the VC has the outside option of investing in a new round of start-ups. This
implies that the VC obtains a surplus equal to l̂2V (SS) =2(3+�)��
6 + v23 , leaving l̂
2E(SS) =
(3��)��6 � v2
6
21The amount of new start-up speci�c investment required from the VC will depend, in general, on the transferability
of the human capital that the VC has accumulated in the past into new start-ups. This implies that the start-up speci�c
investment may change over time, with later start-ups requiring possibly a lower cost.
24
to each entrepreneur. Finally, if both start-ups fail, state FF , the VC will invest in a new round of two
start-ups.
Each entrepreneur chooses her e¤ort level pi to maximize her expected pro�t �̂2DEi , that is
maxpi
�̂2DEi � pipj((3� �)��
6� v26) + pi(1� pj)
(1 + �)��� v22
� p2i2: (27)
The Nash-equilibrium e¤ort p̂2� is
p̂2� � 3((1 + �)��� v2)2(3 + 2���� v2)
: (28)
Stationarity implies that the VC�s expected pro�t from holding a large portfolio, v2, is implicitly de�ned
by
v2 = �h�p̂2��2(l̂2V C(SS) + �v2) + 2p̂
2�(1� p̂2�)(l̂2V C(SF ) + �v2) + (1� p̂2�)2v2i� c: (29)
The following proposition characterizes the VC�s optimal portfolio size.
Proposition 9 Let � � �c and c � �cD(�) (de�ned in the appendix). There are critical values f�Dc ;�D1 ;�D2 ; ��D2 g
such that the VC�s optimal size is as follows:
i) ) for low project payo¤ (0 � � < �D1 ) the VC does not invest in any start-up, �� = 0;
ii) for high project payo¤ (� � ��D2 ) the VC invests in one start-up only, �� = 1;
iii) for moderate project payo¤ and high focus (�D1 � � < �D2 and �Dc � � � 1) the VC invests in two
start-ups, �� = 2.
Proposition 9 shows that the main results of the static setting extend to the dynamic setting as well.
In the dynamic setting, the VC still chooses the size of his portfolio by trading o¤ the rent extraction,
the resource allocation, and the value dilution e¤ect. However, the VC�s ability to start new projects
a¤ects the magnitude of the rent extraction and the resource reallocation e¤ects. First consider the rent
extraction e¤ect. The VC�s ability to start new projects gives the VC an outside option and thus increases
his rent extraction ability (both in a small and a large portfolio). However, the marginal increase in his
rent extraction ability is greater with a small portfolio than with a large portfolio since in the static model
with a small portfolio the VC has no outside options at all. The additional rent extraction ability, however,
comes at the cost of worsening entrepreneurial incentives.
Second consider the resource reallocation e¤ect which refers to the VC�s ability to reallocate his hu-
man capital and resources from one start-up to another. In the static setting, this e¤ect increases the
25
attractiveness of a large portfolio relative to a small portfolio since when the VC has only one start-up he
cannot reallocate resources. In the dynamic setting, however, the ability to start new investments allows
the VC to reallocate resources to new start-ups even when the VC has a small portfolio. Speci�cally, the
VC cannot reallocate his resources across start-ups at a given time, but he can reallocate his resources
over time to new start-ups if the current start-up in his portfolio fails or is completed. Hence, the dynamic
setting increases the desirability of small portfolios relative to large portfolios, especially for start-ups with
moderate fundamentals and high failure rates.
Finally, note that moving from the static to dynamic setting does not a¤ect the value dilution e¤ect
in any way, and this e¤ect continues to favor small portfolios in the dynamic setting.
Since all previously identi�ed e¤ects and trade-o¤s between holding small and large portfolios survive
in the dynamic setting, all our previous results go through under the dynamic speci�cation. The only
di¤erence is that the dynamic setting may result in an increase or a decrease in the desirability of small
portfolios relative to large portfolios depending on parameter values (that is, ��D2>< �2).
6 Empirical implications
The empirical predictions of our model hinge on the factors that a¤ect the value of the critical parameters
in our model, especially � and �.
(i) VCs prefer to limit their portfolio size for start-ups with ex-ante higher quality. High quality
start-ups with strong fundamentals are expected to generate greater value if successful, and therefore are
characterized by a greater value of �. Our model shows that the VC �nds it more desirable to limit his
portfolio size for such start-ups because doing so leads to strong entrepreneurial incentives, which proves
to be most desirable for start-ups with strong fundamentals. Stronger entrepreneurial incentives improve
start-up success probability and, therefore, generate a superior portfolio performance and pro�ts. Thus,
this prediction helps explain the �nding in Kaplan and Schoar (2005) that better performing VC funds
may choose to stay small.
(ii) VCs investing in high-risk technologies manage larger and more focused portfolios. In our analysis,
VCs bene�t from investing in related start-ups (high �) because focus allows a more e¢ cient reallocation
of human capital from one start-up to another. Reallocation of human capital is more likely when the
(endogenous) failure probability for start-ups is relatively high, which happens when� is small or k is large.
A greater level of focus reduces the ex-post ine¢ ciency associated with spreading the VC�s resources across
26
several start-ups, and increases the bene�ts of ex-post resource reallocation. This implies that focused
portfolios are more desirable (all else equal) for start-ups that invest in technologies with high uncertainty
and failure rates. This prediction is consistent with the �ndings of Cumming (2006). This paper documents
that large portfolios are more likely to be observed for VCs investing in life sciences (rather than other
high-tech �rms), and argues that this strategy emerges because there are greater complementarities (i.e.,
higher focus) across entrepreneurial �rms in the life sciences industry.
(iii) VCs expand their portfolios only if they have access to investment opportunities in the sector of
their specialization. In our model, forming a portfolio with a high degree of focus bene�ts the VC in two
ways. First, focus allows the VC to extract more surplus from his start-ups and, second, it facilitates the
reallocation of human capital and resources among start-ups. In both cases, managing a focused portfolio
increases the VC�s investment incentives and rewards his human capital acquisition. This implies that
VCs refrain from investing in start-ups that are not related to their area of specialization, and that they
are more likely to increase their investments when their industry of specialization experiences a positive
technological shock (an increase in investment opportunities). This prediction is consistent with the
�ndings of Gompers, Kovner, Lerner and Scharfstein (2004) which document that venture capital �rms
with the most industry speci�c human capital and experience react most to an increase in investment
opportunities in the sectors of their specialization. The explanation they o¤er for this evidence is that it
is more di¢ cult for diversi�ed and less specialized VCs to redeploy their human capital from the sectors
of their current investment to the sector experiencing an increase in investment opportunities.
(iv) An increase in the availability of VC �nancing increases entrepreneurial e¤ort and leads to better
success rates for start-ups. In Section 5.1 we show that availability of ex-post VC �nancing increases the
entrepreneurs�ability to extract rents from the VC and, hence, their e¤ort. A greater level of entrepreneur-
ial e¤ort leads to a higher success probability for start-ups. Thus, greater availability of VC �nancing, and
a greater ability of entrepreneurs to transfer their start-ups to alternative VCs, are positively correlated
with better success rates for start-ups. This, to our knowledge, a new and testable implication.
(v) In markets with more abundant supply of VCs (relative to entrepreneurs), VCs will hold smaller
portfolios. In markets characterized by a more abundant supply of VCs, competition for start-ups allows
entrepreneurs to extract greater rents from VCs. As a result, VCs�desire for larger portfolios decreases.
Hence, greater competition among VCs leads to smaller portfolios, a new empirical prediction.
(vi) Industry clustering of smaller entrepreneurial businesses will increase VCs willingness to provide
�nancing to such businesses and will have a positive impact on the creation and the development of new
27
businesses. Our analysis shows that smaller size entrepreneurial projects characterized by moderate � will
become �nancially viable only if the VC can combine them in his portfolio and create a positive externality
between them (given by the reallocation e¤ect). Hence, for a small size start-up, existence of competing but
related businesses in the same industry does not necessarily represent a threat, but rather it contributes to
the development and commercialization of the business by increasing VCs�willingness to provide �nancing
for it. This is a novel prediction of our model with the policy implication that encouraging small business
development in similar industries or geographical locations will increase prospects of such businesses to
receive VC funding.
(vii) VCs with the ability and the experience to create synergies across start-ups will hold larger port-
folios. Our model establishes that VCs with a better ability at reallocating resources from one start-up
to another are more willing to hold larger portfolios. One interpretation of this result is that experienced
VCs will be better at reshu ing resources and generating synergies across start-ups and hence will have
a greater willingness to hold larger portfolios.
7 Conclusions
This paper studies the size and focus of a VC�s portfolio. We have identi�ed three main e¤ects of portfolio
size on the VC�s and entrepreneurs� incentives. The �rst one is the rent extraction e¤ect: The VC can
extract higher rents in a larger portfolio by using his ability to reallocate his limited resources from one
start-up to another. This e¤ect, everything else constant, leads to stronger incentives for the VC and
weaker incentives for the entrepreneurs. The second e¤ect is the resource allocation e¤ect: The VC
bene�ts from investing in a large number of start-ups because this increases (all else equal) the probability
that at least one of the start-ups will be successful and thus the VC will have greater chances to earn a
return from his investment. This e¤ect depends on the VC�s ability to reallocate ex-post his resources
from one start-up to another, after observing whether they have been successful or not. This e¤ect has a
positive impact both on the VC�s investment incentives and entrepreneurial incentives. The third e¤ect
is the value dilution e¤ect: A larger portfolio requires the VC to spread his limited resources across a
large number of start-ups, diluting his value-adding role, with a negative impact on both the VC�s and
entrepreneurial incentives.
Our paper has several implications for VC portfolio management. One key message is that limiting
portfolio size may prove to be bene�cial for a VC despite his ability to add a large number of start-ups
28
to his portfolio. This result originates from the fact that VC human capital is a scarce resource and
committing it to a fewer number of start-ups results in stronger entrepreneurial incentives. In addition,
limiting portfolio size is most desirable for start-ups with strong ex-ante fundamentals since improving
entrepreneurial incentives is most valuable for such start-ups. A larger portfolio becomes optimal when
start-up fundamentals become moderate and when the VC can form a focused portfolio. A high level
of focus increases the bene�ts of a large portfolio in terms of the VC�s rent extraction and resource
reallocation ability. We also show that entrepreneurs with smaller businesses may bene�t from belonging
to a large portfolio, rather than a small one, even if this means that the VC can extract more surplus
from them. This happens when a large portfolio is the only way to enable the VC to make the start-up
speci�c investments necessary for the success of the start-ups in his portfolio. In addition, clustering of
smaller size businesses with a common industry focus increases their prospects of obtaining VC funding
and proves to be bene�cial for both entrepreneurs and VCs. Finally, we show that VCs can create value
by engaging in portfolio management, a real life practice employed by many VCs. Portfolio management
refers to early divestitures of some start-ups to extract higher surplus from the remaining start-ups. We
�nd that the VC bene�ts from portfolio management when the relatedness of the start-ups in his portfolio
is high. Under certain conditions, the ability to engage in portfolio management proves to be socially
e¢ cient since it enlarges the set of economically viable start-ups.
29
References
[1] Bernile, G. and E. Lyandres (2003), The optimal size of a venture capitalist�s portfolio, University of
Rochester, working paper.
[2] Binmore, K., A. Rubinstein, and A. Wolinski (1986), The Nash bargaining solution in economic
modelling, Rand Journal of Economics, Vol. 17, p. 176-188.
[3] Casamatta, C. (2003), Financing and advising: Optimal �nancial contracts with venture capitalists,
Journal of Finance, Vol. 58, p. 2059-2086.
[4] Chemla, G., M. Habib, and A. Ljungqvist (2006), An analysis of shareholder agreements, Journal of
the European Economic Association, Vol. 4, p. 1542-4766
[5] Cumming, D. (2006), The determinants of venture capital portfolio size: Empirical evidence, Journal
of Business, Vol. 79, p. 1083-1126.
[6] Gertner, R., D. Scharfstein and J. Stein (1994), Internal versus External Capital Markets, Quarterly
Journal of Economics, p. 1211-1230.
[7] Gompers, P. and J. Lerner (1996), The use of covenants: An empirical analysis of venture partnership
agreements, Journal of Law and Economics, Vol. 39, p. 463-498.
[8] Gompers, P., Kovner, A. Lerner, J and D. Scharfstein (2004), Venture Capital Investment Cycles:
The Role of Experience and Specialization, HBS, working paper.
[9] Grossman, S. and O. Hart (1986), The costs and bene�ts of ownership: A theory of vertical and
lateral integration, Journal of Political Economy, Vol. 94, p. 691-719 .
[10] Hart, O. and J. Moore (1990), Property rights and the nature of the �rm, Journal of Political Economy,
Vol. 98, p. 1119-1158 .
[11] Inderst, R., H. Muller and F. Muennich (2006), The bene�ts of shallow pockets, LSE, working paper.
[12] Kanniainen, V. and C. Keuschnigg (2003), The optimal portfolio of start-up �rms in venture capital
�nance, Journal of Corporate Finance, Vol. 9, Number 5, p. 521-534.
[13] Kaplan, S. and A. Schoar (2005), Private equity performance: Returns, persistence and capital �ows,
Journal of Finance, Vol. 60, p. 1791-1823.
30
[14] Kaplan, S. and P. Stromberg (2003), Financial contracting theory meets the real world: An empirical
analysis of venture capital contracts, Review of Economic Studies, Vol. 70, p. 281-315.
[15] Michelacci, C. and J. Suarez (2004), Business creation and the stock market, Review of Economic
Studies, Vol. 71, p. 459-481.
[16] Noldeke, G. and K. Schmidt (1998), Sequential investments and options to own, Rand Journal of
Economics, Vol. 29, p. 633-53.
[17] Repullo, R. and J. Suarez (2004), Venture capital �nance: A security design approach, The Review
of Finance, Vol. 8, p. 75-108.
[18] Sahlman, W. (1990), The structure and governance of venture capital organizations, Journal of Fi-
nancial Economics, Vol. 27, p. 473-521.
[19] Stole, L. and Z. H. Zwiebel, (1996a), Intra�rm bargaining under nonbinding contracts, Review of
Economic Studies, Vol. 98, p. 375-1158.
[20] Stole, L. and Z. H. Zwiebel, (1996b), Organizational design and technology choice under intra�rm
bargaining, American Economic Review, Vol. 86, p. 195-222.
31
Appendix
Proof of Proposition 1. The �rst order condition of (1) with respect to p is � = kp, which, if solved
for p, gives (3). Substituting (3) into the entrepreneur�s and the VC�s objective functions, given by (1)
and (2) respectively, gives (4).
Proof of Lemma 1. The VC will make the investment if and only if his expected pro�ts are
nonnegative. It is straightforward to see that the VC�s pro�ts, given in (4), are nonnegative if and only if
� � �m =pck:
Proof of Proposition 2. Since the reaction functions of the two entrepreneurs are symmetric, the
Nash-equilibrium of the e¤ort choice subgame is obtained by setting pj � pi in the �rst-order condition
(10), and then solving for pi; giving (11). Substituting the Nash-equilibrium level of e¤ort (11) into the
entrepreneurs�objective function, (8), and in the VC�s objective function, (9), gives the VC�s pro�ts (12)
and the entrepreneurs�pro�ts (13).
Proof of Lemma 2. Direct comparison of p1�with p2�reveals that p1� > p2� if and only if
2
k>
1 + �
k + 2��3
;
which is always true for � � 1: From (3) and (11) we have that
p1� � p2� = �3k(1� �) + 4��2k (3k + 2��)
: (30)
Di¤erentiating (30) with respect to � gives that
@(p1� � p2�)@�
=9k2(1� �) + 8�� (3k +��)
2k (3k + 2��)2 > 0:
Di¤erentiating (30) with respect to � gives that
@(p1� � p2�)@�
= �� 3 (3k � 2�)2 (3k + 2��)
2 < 0;
since p1�(1) < 1 implies that � < k. Di¤erentiating (30) with respect to k yields that
@(p1� � p2�)@k
= ��9k2 (1� �) + 8�� (3k +��)
2k2 (3k + 2��)2 < 0:
Proof of Proposition 3. Consider �rst the case in which the VC selects only one start-up, � = 1: In this
case, from Lemma 1, we have that the VC earns positive expected pro�ts if and only if � � �m �pck:
Consider now the case in which the VC selects a portfolio with two start-ups, � = 2: De�ne �0(�; k)
implicitly by setting �2�V = 0 in (12). It is straightforward to show that @�2�V@� > 0 and @�2�V
@� > 0: Thus,
32
by the implicit function theorem, we have that @�0(�;k)@� < 0: In addition, from @�2�V
@� > 0 it follows that
the VC earns positive expected pro�ts only if and only if � > �0(�; k). Part (i) of the proposition is
obtained by setting �1(�; k) � minf�m;�0(�; k)g: Note that @�1(�;k)@� � 0 since @�0(�;k)
@� < 0 and �m is
independent of �: De�ne now �c as the unique solution to �2�V C = 0 at � = �m, and note that at � = �c
we have that �2�V = �1�V = 0: Let now �c � � < 1: In this case, if �1 � � < �m, the VC earns positive
expected pro�ts if he invests in two start-ups (since � � �1), but he earns negative expected pro�ts if
he invests in one start-up only (since � < �m). Thus, the VC optimally selects two start-ups. Let now
� � �m. In this case, the VC earns positive expected pro�ts both with one or two start-ups, and both
choices are feasible. Therefore, the VC compares his expected pro�ts with one start-up only, �1�V , with his
expected pro�ts with two start-ups, �2�V to decide whether to invest in one or two start-ups. Using (4)
and (12), we have that the VC invests in one start-up if and only if
�2
k>6 (3k +��) (1 + �)
2�2
4 (3k + 2��)2 : (31)
Rearranging the inequality, we obtain that �1�V > �2�V if and only if
P1 � 64�2�2 + 24k� (7� �(2� �))�� 72k2 (�(2 + �)� 1) > 0:
Note that P1 is a convex parabola in � with two roots, ��1and ��2, given by
��1 � 3k
16�
�� (2 + �)� 7� (1 + �)
q�2 + 2�+ 17
�;
��2 � 3k
16�
�� (2 + �)� 7 + (1 + �)
q�2 + 2�+ 17
�:
It is straightforward to show that ��1 < 0 for all 0 < � < 1; k > 0: We also have that ��2 > 0 for � > �c:
This can be seen by noting that at � = �c we have that ��2(�M ; k) = ��1(�c; k) = �m > 0 and
@��2@�
=3kH(�)
16�2p�2 + 2�+ 17
>3k�7p17� 17
�16�2
p�2 + 2�+ 17
> 0;
since H(�) � ���2 + �� 1
�� 17 +
��2 + 7
�p�2 + 2�+ 17 is increasing in �: Therefore, �1�V > �2�V for
� > ��2 and �1�V � �2�V for �m � � < ��2. De�ne then �2(�; k) � ��2(�; k) for �c � � < 1. Finally, note
that ��2(�M ; k) = ��1(�
M ; k) = �m > 0 at � = �c. Thus, for 0 � � < �c, � � �m implies � > ��2 since
@��2
@� > 0 and, hence, that �1V > �2V . De�ne then �2(�; k) � �m for 0 � � < �c, obtaining (ii) and (iii).
Note that we also have shown that @��2
@� > 0 which implies that @�2
@� � 0: Finally, it is straightforward to
see that ��2 is a linear function of k and positive for � > �c. This implies that@�2
@k > 0 for all � > �c
concluding the proof.
33
Proof of Proposition 4. By direct calculation, comparing the VC�s payo¤ from bilateral bargaining
l2BV (SS), given by (14), and that from multilateral bargaining l2V (SS); given by (7), reveals that l2V (SS) �
l2BV (SS) if and only if � � 35 :
Proof of Proposition 5. The �rst-order condition of (16) is
pBi (pj) = (1 + �)4� 3pj8k
� : (32)
Since the reaction functions of the two entrepreneurs are symmetric, the Nash-equilibrium of the e¤ort
choice subgame is obtained by setting pj = pBi in the �rst-order condition (32) and solving for pBi ; giving
(18). Substituting the Nash-equilibrium level of e¤ort (18) into the entrepreneurs�objective function (16)
and VC�s objective function, (17), gives (19) and (20).
Proof of Proposition 6. The proof of this proposition is similar to the proof of Proposition 3. From
(19) de�ne �B1 (�; k) implicitly by setting �2B�V = 0 in (19). It is straightforward to show that @�2B�V
@� > 0
and @�2B�V
@� > 0. Thus, by the implicit function theorem, we have that @�B1 (�;k;c)@� < 0. Also, from @�2B�V
@� > 0,
the VC earns positive expected pro�ts only if � > �B1 (�; k). De�ne � such that �2B�V = �1�V at � = 35 .
Thus, from (4) there is a �c(k) such that we have �m < � for c < �c(k). Since @�B1 (�;k;c)@� < 0 it follows
that �B1 (�; k) < �m for all �c � � � 1. Thus, for 0 � � < �B1 (�; k), we have that �2B�V < 0 and �1�V < 0,
giving part (i) of the proposition. For �B1 (�; k) � � < �m, we have that �2B�V � 0 but �1�V < 0. Thus
the VC optimally invests in two start-ups, �� = 2: For � � �m we have that �2B�V � 0 and �1�V � 0; and
investment in both one or two start-ups is feasible. Therefore, the VC compares his expected pro�ts with
one start-up only, �1�V , with his expected pro�ts with two start-ups, �2B�V to decide whether to invest in
one or two start-ups. Using (4) and (19), we have that the VC invests in one start-up if and only if
�2
k>8 (1 + �)
2�2 (4k + (1 + �)�)
(8k + 3(1 + �)�)2 :
Rearranging the inequality results in �1�V > �2B�V if and only if
P2 � 36 (1 + �)2�2 + 8k (1 + �) (10� 2�(1 + �))� + 128k2 (1� �(2� �)) > 0:
Note that P2 is convex in �, and has two roots, ~�B1 and ~�B2 , given by
��B1 � 2k
9(1 + �)
�2� (2 + �)� 10�
pA�;
��B2 � 2k
9(1 + �)
�2� (2 + �)� 10 +
pA�;
where A � (1 + �)2(4� (2 + �) + 28) : It is straightforward to show that ��B1 is always negative. ��B2 is
34
positive for � > �� since for c � �c(k) we have that ��B2 > �m > 0 for all 1 > � > �� and
@��B2@�
=4k
18
(2� (2 + �) + 14)pA+ 4 (1 + �)
4
pA (1 + �)
2 > 0:
Hence it follows that �1�V � �2B�V if � � ��B2 and �1�V < �2B�V if �m � � < ��B2 : De�ne then �
B2 (�; k) �
��B2 (�; k); giving part (ii). Finally note that@�B
2
@� =@��B
2
@� > 0, and @�B1
@� � 0.
Proof of Proposition 7. By direct calculation, it is possible to show that �B2 (�; k) and �2(�; k)
have a unique intersection for �� � � < 1, which we denote b�. Comparing �B2 (�; k) and �2(�; k) at � = 1yields that �B2 (� = 1; k) = 4
9
�p10� 1
�k � �2(� = 1; k) = 3
4
�p5� 1
�k; which in turn implies that
�B2 (�; k) � �2(�; k) if and only if � � b�. From the de�nitions of �B2 (�; k) and �2(�; k); �B2 (�; k) =
�2(�; k) at � = b� implies that �2B�V = �2�V at � = b�; and which, in turn, implies that �B1 (�; k) and�1(�; k) intersect only once at � = b� for �� � � < 1, since �B1 (�; k) and �1(�; k) are de�ned as solutions
to �2B�V � c = 0 and ��V (2)� c = 0, respectively. Calculating �B1 (�; k) and �1(�; k) at � = 1 yields that
�B1 (1; k) < �1(1; k), which in turn implies �B1 (�; k) � �1(�; k) if and only if � � b� given that �B1 (�; k)
and �1(�; k) intersect only once for �� < � < 1.
Proof of Lemma 3. It is immediate to see from ~p1� = (1 + �1)� and from (24) that ~p1� and ~p2� are
increasing in �1:
Proof of Proposition 8. If the VC selects only one start-up, the entrepreneur and the VC bargain
over the transfer t, where the entrepreneur has the outside option �0. Given that the entrepreneur and
the VC have the same bargaining power, they with agree on a transfer t such that ~�1�E + t� �0 = ~�1�V � t.
This implies that t = 12 (~�
1�V � ~�1�E + �0) and that the VC�s payo¤, net of the transfer to the entrepreneur
is equal to
~�1�V � ~�1�V � t = 1
2(~�1� � �0); (33)
where ~�1� � ~�1�V + ~�1�E is the joint surplus created by the entrepreneur and the VC.22 If the VC selects
two start-ups, he will engage in a multilateral bargaining game with the two entrepreneurs similar to the
one we described in the basic model. Note that if bargaining with one entrepreneur breaks down, the VC
bargain with the remaining one, obtaining ~�1�V . This implies that the VC will agree to pay to entrepreneur
i a transfer ti such that ~�2�Ei + ti � �0 = ~�2�V � ti � tj � ~�1�V . Setting ti = tj � t2 and ~�2�Ei = ~�
2�Ej = ~�
2�E , we
obtain that t2 = 13 (~�
2�V � ~�2�E +�0� ~�1�V ), and that the VC�s payo¤, net of the transfer to the entrepreneur,
is equal to
~�2�V � ~�2�V � t2 =1
3~�2�V +
2
3
�~�2�E � �0 �
1
2(~�1� � �0)
�: (34)
22Note that the condition t � 0 requires that ~�1�E � ~�1�V � �0, which we assume to be the case.
35
Let ~�2� � ~�2�V + 2~�2�E be the joint surplus created by the VC and the two entrepreneurs. This proof now
follows a procedure similar to the proof of Proposition 3. Consider �rst the case in which the VC selects
only one start-up , � = 1. In this case, from (33), we have that ~�1�V � 0 for � � ~�m �q
2c3+�0(2��0) .
Consider next the case in which the VC selects two start-ups, � = 2. De�ne ~�0(�) implicitly by setting
~�2�V = 0 in (34). By direct calculation it is easy to verify that @ ~�2�V@� > 0 and @ ~�2�V
@� > 0. Thus, by
the implicit function theorem, we have that @ ~�0(�)@� < 0. In addition, from @ ~�2�V
@� > 0 it follows that the
VC earns positive expected pro�ts for � > ~�0(�). Part (i) of the proposition is obtained by setting
~�1(�) � minf�m; ~�0(�)g. Note that @ ~�1(�)@� � 0 since @ ~�0(�)
@� < 0 and �m is independent of �: De�ne
now ~�c as the unique solution to ~�2�V = 0 at � = ~�m; and note that at � = ~�c we have that ~�
2�V =
~�1�V = 0: Let now ~�c � � � 1. In this case, if ~�1 � � < ~�m; the VC earns positive pro�ts if he invests
in two start-ups (since � � ~�1), but he earns negative pro�ts if he invests in one start-up only (since
� < ~�m). Thus, the VC optimally selects two start-ups. Let now � � ~�m. In this case, the VC earns
positive expected pro�ts both with one or two start-ups, and both choices are feasible. Therefore, the VC
compares his expected pro�ts with one start-up only, ~�1�V , with his expected pro�ts with two start-ups,
~�2�V . From (33) and (34) we obtain that ~�1�V � ~�2�V if and only if
1
2~�1� � ~�2� � �3�0
2;
that is, if and only if
�2F (�; �) � �3�0 + c2
where F (�; �) � 3+�1(2��1)2 � 3 (2(1 + �) + �1) 2(9+2��)(1+�)+�1(2�(1�2�)�3)(12+(�1+8�)�)
2 . By direct di¤erentiation, it
can be easily veri�ed that @F (�;�)@� > 0 and @F (�;�)@� < 0. This implies that there is a ~�2(�; �0) such that
~�1�V � ~�2�V if and only if � � ~�2(�; �0), with@ ~�2(�;�0)
@� > 0 implied by implicit function di¤erentiation.
This implies that for ~�1 � � < ~�2(�; �0) the VC sets �� = 2, and for � � ~�2(�; �0) the VC sets �� = 1,
proving (iii) and (iv). This also implies that @ ~�2(�;�0)
@�E0< 0. Part (ii) of the proposition is easily veri�ed
by noting that ~�2(~�c) = ~�1(~�c) = �m > 0 at � = ~�c. Thus,@ ~�2(�;�0)
@� > 0 implies that for 0 � � < ~�c
and � � �m we have that � > ~�2 and, hence, that ~�1�V > ~�2�V , concluding the proof.
Proof of Proposition 9. This proof follows a procedure similar to the one used in the proof of
Proposition 3. The VC will choose at each date the portfolio that maximizes his continuation payo¤ by
comparing (26) and (29). By implicit function di¤erentiation of (29) it is easy to show that there is a
�c1 > 0 such that if � � �c1 we have that
@v2
@�> 0 and
@v2
@�> 0: (35)
36
This implies that for all � there is a unique �D(�) 2 [0; 1] such that v2(�;�)�<v1(�) as ��<�D(�). De�ne
�Dm such that v1(�Dm) = 0. Let �̂D0 (�) be such that v
2(�; �̂D0 ) = 0, and let �Dc such that �̂
D0 (�
Dc ) = �
Dm.
Note that v1(�Dm) = v2(�Dc ;�Dm) = 0. By the implicit function theorem, (35) implies that @�̂D
0 (�)@� < 0.
De�ne then �D1 (�) � minf�Dm; �̂D0 (�)g. Monotonicity of v2(�;�) and v1(�) in� imply that v2(�; �̂D) <
0 and v1(�) < 0 for � < �D1 (�), and therefore that the VC will not invest in any start-up for � < �D1 (�),
that is, �� = 0. For � � �D1 (�) the VC will invest in either one or two start-ups. Let �̂D(�) be the
inverse of �D(�). If, for some �, �̂D(�) is not single-valued, let ��2(�) = minf� : �D(�) = �g and
�̂2(�) = maxf� : �D(�) = �g. Let now �Max � c2�(1��(1+�)) . From p̂1� � �� � v1
2 ; it is easy to verify
that p̂1�(�Max) = 1, which means that � � �Max. Consider now p̂2�(�;�; �; v2) as de�ned in (28). By
direct di¤erentiation, it is easy to verify that @p̂2�(�)@� > 0 at � = 0. This implies that there is a �c2 > 0
such that @p̂2�(�)@� > 0 for all � 2 [0; �c2 ]. Furthermore, by direct di¤erentiation, it is also easy to verify
that @p̂2�(�)@v2
< 0 and @p̂2�(�)@� > 0. This implies that
p̂2� =3
2
(1 + �)��� v23 + 2���� v2
<3
2
2��
3 + 2��< (�; c) � 3
2
2��Max
3 + 2��Max:
Since (�; 0) = 0, there is a �cD(�) such that p̂2� < 12 for all c 2 [0; �c
D(�)]. By implicit function di¤erentia-
tion of (29) it is easy to verify that in this case @v2@p̂2� > 0. Thus, by substituting p̂
2� = 12 in (29) we obtain
that v2 � �v2, where
�v2 = �
�1
4(4
3��Max +
�v23) +
2�Max + �v24
) +1
4�v2
�� c:
Thus, solving for �v2, we obtain
v2(�Max; �) � �v2 =
�c6(1��(�+1)) +
�c4�(1��(�+1)) � c
1� 712�
:
Similarly, from (26), we obtain that
v1(�Max; �) =c
2� (1� � (1 + �)) :
Comparing �v2 with v1(�Max) we obtain that v2(�Max; 0) < v1(�Max; 0). Thus, there is a there is a �c3 > 0
such that v2(�Max; �) < v1(�Max; �) for all � 2 [0; �c3 ]. De�ne �Dc � minf�c1 ; �c2 ; �c3g. This implies that
v1(�;�) � v2(�) for � � ��D2 � maxf�Dm; �̂2(�)g and that v1(�;�) < v2(�) for �D1 � � < �D2 , where
�D2 � ��2(�), concluding the proof.
37